LRT calculation in tidyLPA

General
1

Hi everyone,

I am trying to calculate Lo-Mendell-Rubin likelihood ratio test (LRT) after Latent Profile Analysis using tidyLPA (the distribution of my indicators are non-normal, so the LRT seems preferable). The LRT is not provided by default, however the documentation says it can be calculated as follows:


calc_lrt(n, null_ll, null_param, null_classes, alt_ll, alt_param, alt_classes)

However, I am uncertain where I can retrieve the parameters that need to be specified in the parentheses and where to insert the calc_lrt()-function in my code.

For example, the main part of my code looks like this:

# estimate profiles 
est.m3.m6.compl_sub1 <- dta2.1[complete.cases(dta2.1$X, dta2.1$Y, dta2.1$Z, dta2.1$V), ] %>% 
  select(c("X", "Y", "Z", "V")) %>% 
  estimate_profiles(1:6, models = c(3, 6)
  ) 
est.m3.m6.compl_sub1

# full model indices
get_fit(est.m3.m6.compl_sub1)

# class estimates
get_estimates(est.m3.m6.compl_sub1)

Thank you so much in advance!

2

This topic was automatically closed 42 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.